Stimulus and recognition thresholds for the basic tastes in deionized ...

Science & Research | Original Contribution

Beleg/Autorenexemplar! Jede Verwertung außerhalb der engen Grenzen des Urheberrechtsgesetzes ist ohne Zustimmung des Verlages unzulässig und strafbar. Dies gilt insbesondere für Vervielfältigungen, Übersetzungen, Mikroverfilmungen und die Einspeicherung und Verarbeitung in elektronische Systeme.

Peer-reviewed | Manuscript received: March 20, 2014 | Revision accepted: May 26, 2014

Stimulus and recognition thresholds for the basic tastes in deionized water Are the recommendations for citric acid too high? Karolin Höhl, Gesa U. Schönberger, Heidelberg; Mechthild Busch-Stockfisch, Hamburg

Summary During the recruitment of sensory panelists (e.g. when screening assessors suitability or training of assessors), the sensitivity for basic tastes is often measured with threshold tests. The present study examined the concentrations of taste stimuli used in the threshold tests in accordance with ISO and DIN [1, 2], with respect to their applicability in deionized water. The taste thresholds for sweet, sour, salty, bitter, umami and metallic were determined for 70 young European women. With eight increasing concentrations, it was found that the individual taste stimuli for sweet, salty, bitter, umami and metallic were well distributed. However, sour was recognized by all study participants within the two lowest concentrations. Clear differences in sensitivity were achieved by reducing the citric acid concentration to a fifth of the original value. Thus, if deionized water is used as solvent, the citric acid concentration must be reduced if the expected differences in perceived sourness are to be found. Keywords: taste sensitivity, threshold, citric acid, type of water, sensory science

Introduction The international standard ISO 3972 [1] and the corresponding national DIN standard [2] provide

Citation: Hoehl K, Schoenberger GU, Busch-Stockfisch M (2014) Stimulus and recognition thresholds for the basic tastes in deionized water. Are the recommendations for citric acid too high? Ernahrungs Umschau 61(8): 130–136 This article is available online: DOI: 10.4455/eu.2014.023

guidelines for examining human taste sensitivity in the context of sensory trials. For example, recommendations are given on the chemicals and reagents to be added, e. g., the type of water used for preparing solutions of taste stimuli. The water should be “neutral, tasteless, still and odourless, preferably of known hardness” [1]. For international taste research, deionized water is mostly used to prepare the solutions (see e.g. [3–6]). This ensures that comparability betwen different study centres is ensured and that the composition of the water is relatively stable – independent of the site of production. Nevertheless, the evaluation of the

130 Ernaehrungs Umschau international | 8/2014

results of a large study [7] raised doubts as to whether deionized water is suited for the measurement of taste sensitivity for sour.1 It is thought that either (a) the citric acid concentration would have to be adjusted or (b) another type of water would have to be used to dissolve the substance. Moreover, ISO 3972 [1] (with the corrigendum of 2012 [8]), as well as the corresponding national DIN ISO 3972 [2] refer – in contrast to the previous versions [9, 10] – to lower thresholds for bitter and sour in demineralised water, without explaining what is meant by “lower” or giving detailed references to alternative concentrations for the substances. The present study therefore investigates the thresholds for the five basic tastes (sweet, sour, salty, bitter and umami), as well as the metallic sensation, in deionized water. In addition, various citric acid concentrations in deionized water were examined with respect to taste sensitivity in the threshold test.

1

In the original study design [7], it was intended to investigate the influence of sensory training on the perception of sweet, sour and bitter. However, the data for the perception of sourness were not evaluated, as the results were unclear.

Taste substance

Chemical formula

Molecular weight (g/mol)

Taste quality

Stock solution (g/L)

Concentration series*

Unit

Sucrosea

C12H22O11

342.3

sweet

24.00

0.98–35.06

mmol/L

Caffeine

C8H10N4O2

194.19

bitter

0.54

0.29–1.39

mmol/L

Sodium chloridec

NaCl

58.44

salty

4.00

2.81–34.22

mol/L

Monosodium glutamate monohydrateb

NaOOCCH2 CH(NH2)COOH (H2O)

187.13

umami

2.00

0.44–5.34

mmol/L

Iron(II) sulfate heptahydrateb

FeSO4 (7 H2O)

278.01

metallic

0.016

0.0024–0.029

mmol/L

Citric acid monohydratea

C6H8O7 (H2O)

210.14

sour

Group A (ISO)

1.20

0.60–2.86

mmol/L

Group B (ISO/5)

0.24

0.12–0.57

mmol/L

Group C (ISO/10)

0.12

0.06–0.29

mmol/L

b

Table 1: Series of taste dilutions for the basic tastes and the metallic sensation, in accordance with [1, 2] Prepared by dilution of the stock solution (eight sequential 10-fold dilutions; D8–D1) Riedel-de Haen, Sigma-Aldrich Laborchemikalien GmbH, Seelze, Germany; b Fluka, Sigma-Aldrich Chemie GmbH, Steinheim, Germany; c Merck KGaA, Darmstadt, Germany *

a

Materials and methods Study participants 70 female students of European origin (mean age = 22.1 years, standard deviation [s] = 2.5; mean BMI = 21.5 kg/m2, s = 2.4) took part in the present project. They were part of a larger study in the Hamburg University for Applied Sciences (HAW) [7].

Study design and groups The study participants were requested to come to the HAW Sensory Science Laboratory (designed in accordance with the guidelines in [11]) on one of three weekdays and at one of four possible times. They were not informed of the study design or objectives. A session lasted about 1.5 hours and included various tests on taste sensitivity: [a] a matching test, to make the study participants familiar with the five basic tastes, and the metallic sensation; [b] an inten-

sity test with one sweet and one bitter sample and [c] two times three threshold tests with a break to allow regeneration of the taste buds. The sessions on the three weekdays only differed in the concentrations used to prepare the sour solutions (citric acid); the other tastes (sweet, salty, bitter, umami and metallic) were tested in all sessions in the concentrations recommended in [1]: - Day 1 (Group A): Original concentration as given in [1]: citric acid stock solution = 1.2 g/L; n = 15; - Day 2 (Group B): A fifth of the original concentration: citric acid stock solution = 0.24 g/L; n = 30; - Day 3 (Group C): A tenth of the original concentration: citric acid stock solution = 0.12 g/L; n = 25.

Taste substances and type of water The substances (• Table 1) were dissolved in deionized water – provi-

ded by Sensient Food Colors GmbH, Geesthacht, Germany. • Table 2 gives the details of the taste solution for citric acid, as prepared by dilution of the three stock solutions (Group A: ISO concentration as given in [1], Group B: ISO/5 and Group C: ISO/10).

Determination of the stimulus and recognition thresholds The stimulus and recognition thresholds for the six taste qualities were measured consecutively (two trays with three series each). For each tray, the study participants were presented with three taste series of nine samples each (in 40 mL plastic cups, each containing 20 mL solution); the first sample always consisted of deionized water. The following eight solutions (D8–D1, • Tables 1 and 2) were tasted in the order of increasing concentrations (method

▸ Ernaehrungs Umschau international | 8/2014 131


Group A (ISO concentration) Stock solution g/L Dilution

Group B (ISO/5)

Group C (ISO/10)

1.20

0.24

0.12

D1

0.60

0.12

0.06

D2

0.48

0.096

0.048

D3

0.384

0.0768

0.0348

D4

0.3072

0.06144

0.03072

D5

0.246

0.0492

0.0246

D6

0.1968

0.04936

0.01968

D7

0.1572

0.03144

0.01572

D8

0.126

0.0252

0.0126

Table 2: Series of taste dilutions for sour taste perception (citric acid) in three study groups

sweet (log [µmol/L])

salty (log [µmol/L])

bitter (log [µmol/L])

umami (log [µmol/L])

metallic (log [µmol/L])

µ

s

µ

s

µ

s

µ

s

µ

s

Group A (n = 15)

3.2

0.2

6.5

0.1

2.5

0.03

2.7

0.1

0.4

0.04

Group B (n = 30)

3.1

0.3

6.6

0.2

2.5

0.1

2.8

0.2

0.5

0.2

Group C (n = 25)

3.2

0.3

6.5

0.1

2.5

0.1

2.7

0.2

0.5

0.2

overall (n = 70)

3.2

0.3

6.5

0.2

2.5

0.1

2.7

0.2

0.5

0.2

Table 3: M eans (μ) and standard deviation (s) for the stimulus thresholds (sweet, salty, bitter, umami and metallic) in (log [μmol/L]) as in the series of taste dilutions recommended in [1] (see • Table 1)

sweet (log [µmol/L])

salty (log [µmol/L])

bitter (log [µmol/L])

umami (log [µmol/L])

metallic (log [µmol/L])

µ

s

n

µ

s

n

µ

s

n

µ

s

n

µ

s

n

Group A (n = 15)

3.7

0.5

14

6.8

0.4

13

2.7

0.2

14

3.3

0.5

11

0.6

0.4

14

Group B (n = 30)

3.8

0.5

30

7.2

0.5

18

2.7

0.3

26

3.1

0.3

29

1.0

0.3

29

Group C (n = 25)

3.7

0.4

24

7.0

0.4

23

2.7

0.2

25

3.1

0.4

23

0.8

0.5

21

overall (n = 70)

3.7

0.5

68

7.1

0.4

54

2.7

0.2

65

3.2

0.4

63

0.8

0.4

64

**

*

F = 4.23; P ≤ 0.019

F = 5.14; P ≤ 0.008

Table 4: M eans (μ) and standard deviation (s) for the recognition thresholds (sweet, salty, bitter, umami and metallic) in (log [μmol/L]) as in the series of taste dilutions recommended in [1] (see • Table 1) *

P ≤ 0,05; ** P ≤ 0,005


80

sweet salty

70

bitter umami metallic

60 50 percent

as given in [1]). Between the series and between the trays, there was an interval that was long enough to allow regeneration of the taste buds. For this purpose, the study participants rinsed with deionized water and were allowed to eat unleavened bread (matzo) if they wanted. The presentation of the series of taste dilutions was randomised between the study participants. The stimulus threshold is defined as the minimum concentration of a sensory stimulus needed to give rise to a sensation (different to the sensation from the initial water sample). However, the sensory sensation must not be identifiable quantitatively. The recognition threshold is the minimum concentration of a sensory stimulus that permits the assessor to make a qualitative recognition of the sensory impression.

40 30 20 10 0

D8

D7

D6

D5

D4

D3

D2

D1

not recognized

Taste dilution

Figure 1: S timulus thresholds for the five tastes (sweet, salty, bitter, umami and metallic) in deionized water (n = 70) as in the series of taste dilutions recommended in [1] (see • Table 1)

sweet

Statistics

salty

35

bitter umami metallic

30 25 percent

The evaluations were performed with SPSS Statistics 21.0. For the calculation of the means (μ) and standard deviation (s), the concentrations (μmol/L) were transformed to logarithmic values (log [μmol/L]). To calculate the group differences, one-way analysis of variance (ANOVA) was performed. The level of significance was taken as a ≤ 0.05.

20 15 10

Results The mean values for the stimulus and recognition thresholds (sweet, salty, bitter, umami and metallic) of the three study groups (A, B and C) are shown in • Tables 3 and 4. If the three groups give the same mean threshold values (all three groups tasted identical concentrations according to [1]), the text gives the group mean values. • Figures 1 and 2 show the distribution of the individual results over the eight threshold concentrations (D8–D1) in ISO concentration for the five tastes. • Table 5 shows the mean stimulus and recognition threshold for the

5 0

D8

D7

D6

D5

D4

D3

Taste dilution

D2

D1

not recognized

Figure 2: R ecognition thresholds for the five tastes (sweet, salty, bitter, umami and metallic) in deionized water (n = 70) as in the series of taste dilutions recommended in [1] (see • Table 1)

sour taste in three different concentrations (Group A: ISO concentration as given in [1]; Group B: ISO/5; Group C: ISO/10). • Figures 3 and 4 show the distribution of the individual results of the threshold tests for the sour sensation

for the three concentration series.

Stimulus and recognition thresholds for sweet, salty, bitter, umami and metallic As expected, the three study groups



s

n

Recognition threshold (log [µmol])

s

n

Group A (ISO-concentration as given in [1]) 2.8

0.00

15

2.8

0.03

15

Group B (ISO/5-concentration)

2.1

0.1

30

2.2

0.1

30

Group C (ISO/10-concentration)

1.8

0.1

25

1.9

0.1

19

Stimulus threshold (log [µmol])

Table 5: M eans (μ) and standard deviations (s) of the stimulus and recognition thresholds for sour (citric acid) for the three study groups

(A, B and C; n = 70) do not differ in their stimulus thresholds with respect to sweet (mean for the whole group [μ] = 3.2 log [μmol/L]; ~D7), salty (μ = 6.5 log [μmol/L]; ~D7), bitter (μ = 2.5 log [μmol/L]; ~D8), umami (μ = 2.7 log [μmol/L]; ~D7) or metallic (μ = 0.5 log [μmol/L]; ~D8/D7; • Table 3), as there were no concentration differences in the taste dilutions between the three study groups. As expected, the recognition thresholds for sweet (μ [n = 68] = 3.7 log [μmol/L]; ~D5/ D4), bitter (μ [n = 65] = 2.7 log [μmol/L]; ~D6/D5) and umami (μ [n = 63] = 3.2 log [μmol/L]; ~D5/ D4) did not differ significantly between the three study groups. Nevertheless, there were significant differences between Groups A and B with respect to the means for the salty and metallic recognition threshold (• Table 4). For all five tastes, the individual results exhibit satisfactory distributions over the eight dilutions (D8-D1; • Figures 1 and 2). At the lowest concentration (D8), about 60–70 % of the study participants perceived a sensation that differed from that of water. Nevertheless, there were also study participants who only perceived a sensation at much higher concentrations (e.g. at D2 for sweet, salty and metallic; • Figure 1). The individual recognition thresholds were even more clearly distributed over the whole range of the dilutions. Moreover, for each taste, there were study participants who

were unable to assign the test quality correctly (e.g. ~23 % for the salty taste; • Figure 2).

Stimulus and recognition thresholds for sour with ISO concentrations At the lowest concentration (D8), all study participants in Group A (ISO concentration) already perceived a sensation that differed from the water sample (stimulus threshold; • Figure 3). The mean stimulus threshold in this group was 2.8 log (μmol/L; • Table 5). In addition, they recognized the sour taste within the two lowest concentrations (D8 and D7; recognition threshold; • Figure 4). Thus, the mean recognition threshold for Group A was just as low, at 2.8 log (μmol/L; • Table 5). This result confirms the finding in [7], that the citric acid concentrations recommended in [1, 2] give rise to little or no differences in the study participants’ taste sensitivity, when the substance is dissolved in deionized water.

Stimulus and recognition thresholds for sour with reduced concentrations As the starting concentrations of the solutions for Group B (ISO/5) and Group C (ISO/10) were much lower than for Group A, the mean thresholds for these two groups were also much lower: The mean ISO/5 recognition threshold for Group B was 2.2 log (μmol/L), which is some-


what above D7; the mean ISO/10 recognition threshold for Group C was 1.9 log [μmol/L], which lies between D7 and D6 (• Table 5). Moreover, reducing the concentrations to one fifth or one tenth of the original enhanced the distribution of the individual taste sensitivities over the different dilutions. • Figure 3 shows the individual results for the stimulus threshold. The stimulus thresholds for ISO/5 and ISO/10 (Groups B and C) were within the first three concentrations (D8–D6). It is evident from • Figure 4 (recognition thresholds), that only 37 % of Group B (ISO/5) recognized the sour taste at the lowest concentration (D8). Four study participants (16 %) in Group C (ISO/10) recognized the sour taste at the lowest concentration (D8). Moreover, when the concentration was reduced to one tenth, six study participants (24 %) did not recognize the sour taste at all.

Discussion Comparison of the individual results (• Figure 1 and 2) gave the expected picture regarding both stimulus and recognition threshold of all five tastes – sweet, salty, bitter, umami and metallic. There were study participants with sensitive taste abilities, who perceived and recognized a taste at very low concentrations. At the same time, there was a distribution over the whole concentration range (D8–D1) and study participants who failed to assign the taste properly (not recognized).

In a follow-up project [14] to the present study, it could be shown that the recommended citric acid concentration (original concentrations as given in [1, 2]) is more suitable when the substance is dissolved in spring water or tap water. Both the stimulus and recognition thresholds for sour differ significantly between the three water types. The highest threshold values for sour were determined in tap

water, with intermediate values in spring water and the lowest values in deionized water. As explained above, the very low mean threshold values in deionized water were due to the lack of differences between the study participants with respect to taste sensitivity. This homogeneity of the individual taste sensitivity was unexpected and is due to the excessively high initial concentration of citric acid when this is dissolved in deionized water. After evaluation of the present study results, it is unclear why there are differences between the study groups with respect to their mean recognition thresholds for salty and metallic, although there were no concentration differences in the tasted solutions. In Group B, there was a high proportion of study participants who totally failed to recognize the salty taste (n = 12; 40 %). This might be due to study participants exhaustion or to errors in sample preparation or presentation (e.g. if the NaCl stock solution was too low or in the false sequence in the sample series). Study participants frequently had problems in

recognizing the metallic sensation [14]. In future studies, it must be checked whether the concentrations should be increased or whether specific training could lead to improved sensitivity for this unfamiliar taste. For the sour taste sensation, it must be concluded that special care is required for the selection of the type of water used to dissolve the citric acid. Spring water with low total mineral content can be appropriately used for all basic tastes aside from metallic [14]. As it is so readily available, it is a suitable solvent for practical work – such as screening candidates for sensory testing or for training sensory abilities. In further studies, it should be examined to what extent filtered tap water (e.g. with an ion exchanger or active carbon) gives comparable results for taste sensitivity. With this procedure, it would be even simpler and cheaper to prepare the taste solutions than with spring water. On the other hand, if deionized water has to be used (e.g. to guarantee international comparability), it should be considered reducing the citric acid concentration for the stock solution to prepare the

100

A: ISO (1.2g/L) B: ISO/5 (0.24g/L)

90

C: ISO/10 (0.12g/L)

80 70 Percent

The situation was different for the sour taste in the ISO concentration (Group A): In this case, 100 % of the study participants detected a sensation at the lowest concentration (D8) and recognized this as sour within the lowest two concentrations (D8 and D7). It can therefore be inferred that the citric acid concentrations recommended in [1, 2] to measure thresholds for taste sensitivity are too high when the substance is dissolved in deionized water. For the other substances, the concentrations in deionized water appear to be more appropriate. This conclusion is confirmed by the results of another study [12], in which the type of water used to dissolve the substances has a greater influence on sour perception than on the perception of other tastes. If the concentration was reduced to one fifth, the distribution of taste sensitivity was improved (Group B). Dilution to a tenth appeard to be too low, as then 24 % of the study participants in Group C failed to recognize the taste as sour. The concentration range of ~120 to ~571 μmol/L (dilution series of one-fifth of ISO [1]) used in our study was comparable to the range of 48–720 μmol/L employed in [13], although the authors did not fully explain its origin. They prepared a series of 15 citric acid dilutions in deionized water (D15–D1; 48–720 μmol/L). The highest concentration in the present study (D1) roughly corresponded to their twelfth highest concentration [13].

60 50 40 30 20 10 0 d

ize t no cogn re

Taste dilution Figure 3: S timulus thresholds for sour (citric acid) in deionized water for the three study groups A (n = 15), B (n = 30) and C (n = 25)


Science & Research | Original Contribution 100

A: ISO (1.2g/L) B: ISO/5 (0.24g/L)

90

C: ISO/10 (0.12g/L)

80 Percent

70 60 50 40 30 20 10 0 d

ize t no cogn re

Taste dilution Figure 4: Recognition threshold for sour (citric acid) in deionized water for the three study groups A (n = 15), B (n = 30) and C (n = 25)

series of taste dilutions (e.g. to one fifth of the concentration recommended in [1]).

Dr. Karolin Höhl1 Dr. Gesa U. Schönberger1 Prof. Dr. Mechthild Busch-Stockfisch2 Dr. Rainer Wild-Stiftung Mittelgewannweg 10, 69123 Heidelberg 2 Hochschule für Angewandte Wissenschaften Hamburg, Fakultät Life Sciences, Fachbereich Ökotrophologie, Lohbrügger Kirchstraße 65, 21033 Hamburg

1

Korrespondenz: Dr. Karolin Höhl; E-Mail: [email protected]

Conflict of Interest M. Busch-Stockfisch is charmain of the DIN Standards Comitee for Sensory Sciencie and German ISO Delegate for Sensory Sciencie. K. Hohl und G.U. Schönberger declare no conflict of interest according to the guidelines of the International Committee of Medical Journal Editors.

References 1. International Organization for Standardization. ISO 3972. Sensory analysis - Methodology - Method of investigating sensitivity of

taste. ISO, Geneva/Switzerland (2011) 2. Deutsches Institut für Normung. DIN ISO 3972. Sensorische Analyse - Methodologie Bestimmung der Geschmacksempfindlichkeit. DIN e.V., Berlin (2013) 3. Frank RA, Archambo G (1986) Intensity and hedonic judgments of taste mixtures: an information integration analysis. Chem Senses 11: 427–438 4. Algom D, Marks LE (1989) Memory psychophysics for taste. Bull Psych Soc 27: 257–259 5. Pickering GJ, Simunkova K, DiBattista D (2004) Intensity of taste and astringency sensations elicited by red wines is associated with sensitivity to PROP (6-n-propylthiouracil). Food Qual Pref 15: 147–154 6. Nachtsheim R, Schlich E (2014) The influence of oral phenotypic markers and fat perception on fat intake during a breakfast buffet and in a 4-day food record. Food Qual Pref 32: 173–183 7. Hoehl K, Schoenberger GU, Schwarz K et al. (2013) Is perception of sucrose and caffeine affected by training or experience? Monitoring training effects in female subjects over a halfyear period. J Sensory Studies 28: 1–13 8. International Organization for Standardization. ISO 3972. Sensory analysis - Methodology - Method of investigating sensitivity of taste. ISO, Geneva/Switzerland (2011/Cor 1-2012) 9. International Organization for Standardization. ISO 3972. Sensory analysis - Methodology - Method of investigating sensitivity of


taste. ISO, Geneva/Switzerland (1991) 10. Deutsches Institut für Normung. DIN 10959. Sensorische Prüfverfahren – Bestimmung der Geschmacksempfindlichkeit. DIN e.V., Berlin (2005) 11. International Organization for Standardization. ISO 8589. Sensory analysis – General guidance for the design of test rooms. ISO, Geneva/Switzerland (2007) 12. Hough G, Martinez E, Contarini A et al. (1995) Selection of assessors based on their skill in identifying basic tastes in low concentration solutions. J Sensory Studies 10 (1): 1–14 13. Gomez FE, Cassís-Nosthas L, Morales-deLeón JC et al. (2004) Detection and recognition thresholds to the 4 basic tastes in Mexican patients with primary Sjögren’s syndrome. Eur J Clin Nutrition 58 (4): 629–636 14. H oehl K, Schoenberger GU, Busch-Stockfisch M (2010) Water quality and taste sensitivity for basic tastes and metallic sensation. Food Qual Pref 21: 243–249

DOI: 10.4455/eu.2014.023